Back to EveryPatent.com
United States Patent |
6,165,744
|
Kinet
,   et al.
|
December 26, 2000
|
Isolation and characterization of cDNAs coding for the .alpha., .beta.,
and .gamma. subunits of the high-affinity receptor for immunoglobulin E
Abstract
The present invention relates to DNA segments encoding the .alpha., .beta.,
and .gamma. subunits of the high affinity receptor for immunoglobulin E
(IgE). The invention further relates to a method of producing the receptor
by expressing cDNA for its .alpha., .beta., and .gamma. subunits in a host
cell simultaneously.
Inventors:
|
Kinet; Jean-Pierre (Bethesda, MD);
Metzger; Henry (Chevy Chase, MD)
|
Assignee:
|
The United States of America as represented by the Department of Health (Washington, DC)
|
Appl. No.:
|
626704 |
Filed:
|
December 14, 1990 |
Current U.S. Class: |
435/69.1; 435/252.3; 435/320.1; 530/350; 536/23.5 |
Intern'l Class: |
C07K 014/705; C12N 015/12 |
Field of Search: |
435/69.1,252.3,320.1
536/27,23.5
530/350
|
References Cited
Other References
Am. Rev. Immunol. 4:419-430, 1986, Metzger et al. The Receptor with High
Affinity for Immunoglobulin E.
Nature 313:806-809, Feb. 28, 1985, Jacoby et al. Isolation and
Characterization of Genome and cDNA Clones of Human Erythropoietic.
Biochem. 24: 4117-24, 1985, Kinet et al. Dissociation of the Receptor for
Immunoglobulin E in Mild Detergents.
Biochemistry 24:7342-48, 1985, Kinet et al. Noncovalently and Covalently
Bound Lipid on the Receptor for Immunoglobulin E.
J. Biol. Chem. 259:14922-27, Alcaraz et al. Phase Separation of the
Receptor for Immunoglobulin E and Its Subunits in Triton X-114.
|
Primary Examiner: Ulm; John
Attorney, Agent or Firm: Klarquist Sparkman Campbell Leigh & Whinston,LLP
Parent Case Text
This application is a continuation in part of application Ser. No.
07/259,065, filed Oct. 18, 1988 abandoned, which is a continuation-in-part
of U.S. patent application Ser. No. 07/160,457, filed Feb. 24, 1988 now
U.S. Pat. No. 5,639,660, and U.S. patent application Ser. No. 07/240,692,
filed Sep. 6, 1988 abandoned, the contents of all of which are
incorporated herein by reference.
Claims
What is claimed:
1. An isolated nucleic acid molecule encoding a beta subunit of rat
Fc.sub..epsilon. RI, said beta subunit having an amino acid sequence of at
least amino acid residues 1-107, as shown in FIG. 6A.
2. The nucleic acid molecule of claim 1 wherein said molecule comprises at
least nucleotides 55-375 of a nucleotide sequence as shown in FIG. 6A.
3. The nucleic acid molecule of claim 1, wherein said amino acid sequence
comprises at least amino acid residues 1-243, as shown in FIG. 6A.
4. The nucleic acid molecule of claim 3, wherein said nucleic acid molecule
comprises at least nucleotides 55-783 of a nucleotide sequence as shown in
FIG. 6A.
5. A recombinant vector including a nucleic acid molecule according to
claim 4.
6. A transgenic cell produced by introducing into a cell a vector according
to claim 5.
7. A recombinant vector including a nucleic acid molecule according to
claim 3.
8. A transgenic cell produced by introducing into a cell a vector according
to claim 7.
9. A method of expressing a functional rat Fc.sub..epsilon. RI in a host
cell, comprising introducing into a host cell capable of expressing
.alpha. and .gamma. subunits of rat Fc.sub..epsilon. RI a nucleic acid
molecule according to claim 3, and culturing the cell under conditions
whereby a functional rat Fc.sub..epsilon. RI is expressed.
10. A method of expressing a functional rat Fc.sub..epsilon. RI in a host
cell, comprising introducing into the host cell nucleic acid molecules
encoding
(a) an .alpha. subunit of rat Fc.sub..epsilon. RI;
(b) a .beta. subunit of rat Fc.sub..epsilon. RI; and
(c) a .gamma. subunit of rat Fc.sub..epsilon. RI wherein said .beta.
subunit is encoded by a nucleic acid molecule according to claim 3.
11. The nucleic acid molecule of claim 1, wherein said amino acid sequence
comprises amino acid residues 1-107 of FIG. 6A followed by amino acid
residues 1-5 of an amino acid sequence shown in FIG. 6B.
12. A recombinant vector including a nucleic acid molecule according to
claim 11.
13. A transgenic cell produced by introducing into a cell a vector
according to claim 12.
14. A recombinant vector including a nucleic acid molecule according to
claim 1.
15. A transgenic cell produced by introducing into a cell a vector
according to claim 14.
16. A cell according to claim1 15 wherein the cell is a bacterial cell.
17. A cell according to claim 15 wherein the cell is a eukaryotic cell.
18. A cell according to claim 17 wherein the cell is a mammalian cell.
19. A method of producing a beta subunit of rat Fc.sub..epsilon. RI, the
method comprising growing a cell according to claim 15 under conditions
whereby the nucleic acid molecule is expressed, resulting in the synthesis
of said beta subunit in the cell.
20. The method of claim 19 further comprising purifying the beta subunit
from the cell.
21. An isolated nucleic acid molecule including at least 15 contiguous
nucleotides of the nucleotides 55-375 shown in FIG. 6A, or at least 15
contiguous nucleotides of a complement of nucleotides 55-375.
22. A recombinant vector including a nucleic acid molecule according to
claim 21.
23. A transgenic cell produced by introducing into a cell a vector
according to claim 22.
24. A recombinant vector including a nucleic acid molecule according to
claim 21.
25. A transgenic cell produced by introducing into a cell a vector
according to claim 24.
26. An isolated nucleic acid molecule including at least 18 contiguous
nucleotides of a DNA sequence shown in FIG. 6A, or the complement of said
sequence.
27. The isolated nucleic acid molecule of claim 26, including at least 20
contiguous nucleotides of the DNA sequence shown in FIG. 6A, or the
complement of said sequence.
28. The isolated nucleic acid molecule of claim 26, including at least 26
contiguous nucleotides of the DNA sequence shown in FIG. 6A, or the
complement of said sequence.
Description
BACKGROUND OF THE INVENTION
FIELD OF THE INVENTION
The present invention relates to DNA segments encoding the .alpha., .beta.,
and .gamma. subunits of the high affinity receptor for immunoglobulin E
(IgE). The invention further relates to a method of producing the receptor
by expressing DNA encoding its .alpha., .beta., and .gamma. subunits in a
host cell simultaneously.
BACKGROUND OF THE INVENTION
Receptors that bind the Fc region of immunoglobulins ("Fc receptors")
mediate immunoglobulin transport across membranes, stimulate a variety of
cellular activities induced by antigen-antibody complexes, and possibly
regulate the biosynthesis of antibodies. Three of the receptors (the
receptor for polymeric immunoglobulin (Mostov et al. (1984) Nature
(London) 308:37-43), the Fc receptors on macrophages and lymphocytes
(Ravetch et al. (1986) Science 234:718-725), and the high-affinity Fc,
receptor on mast cells and basophils (Kinet et al. (1987) Biochemistry
26:4605-4610; Shimizu et al. (1988) Proc. Natl. Acad. Sci. USA
85:1907-1911; Kochan et al. (1988) Nucleic Acids Res. 16:3584) share a
common feature: their immunoglobulin-binding portion contains two or more
immunoglobulin-like domains.
The receptor with high affinity for IgE (Fc.sub..epsilon. RI) is found
exclusively on mast cells, basophils and related cells. Aggregation of IgE
occupied Fc.sub..epsilon. RI by antigen triggers both the release of
preformed mediators such as histamine and serotonin, as well as
stimulation of the synthesis of leukotrienes. It is the release of these
mediators that results in the allergic condition.
The most thoroughly characterized Fc.sub..epsilon. RI is that of the rat
basophilic leukemia (FEL) cell line. It consists of three different
subunits: (1) a 40-50 Kilodalton (Kd) glycoprotein alpha chain which
contains the binding site for IgE, (2) a single 33 Kd beta chain and (3)
two 7-9 Kd disulfide linked gamma chains (H. Metzger et al, Ann. Rev.
Immunol. 4:419-470 (1986)).
Complementary DNA (cDNA) for the rat .alpha. subunit has recently been
isolated (J.-P. Kinet et al, Biochemistry 26:4605-4610 (1987)). However,
previously there has been no disclosure of the isolation and
characterization of the .beta. and .gamma. subunits or of a cDNA of the
human .alpha. subunit; nor has it been possible to express IgE-binding by
transfected cells (J.-P. Kinet et al, Biochemistry 26:4605-4610 (1987); A.
Shimizu et al, Proc. Natl. Acad. Sci. USA 85:1907-1911 (1988)).
The present invention provides in one embodiment, a cDNA clone for the
alpha subunit of the human Fc.sub..epsilon. RI. The invention further
provides, in further embodiments, cDNA clones for the .beta. and .gamma.
subunits of Fc.sub..epsilon. RI. The invention still further provides a
method of producing the Fc.sub..epsilon. RI receptor.
SUMMARY OF THE INVENTION
It is a general object of this invention to provide DNA segments encoding
Fc.sub..epsilon. RI.
It is a specific object of this invention to provide DNA segments of the
.alpha., .beta., and .gamma. subunits of Fc.sub..epsilon. RI.
It is a further object of the invention to provide polypeptides
corresponding to the .alpha., .beta., and .gamma. subunits of
Fc.sub..epsilon. RI.
It is another object of the invention to provide a recombinant DNA molecule
comprising a vector and a DNA segment encoding the .alpha., .beta., or
.gamma. subunits of Fc.sub..epsilon. RI.
It is a further object of the invention to provide a cell that contains the
above-described recombinant DNA molecule.
It is another object of the invention to provide a method of producing
polypeptides having amino acid sequences corresponding to the .alpha.,
.beta., and .gamma. subunits of Fc.sub..epsilon. RI.
It is a further object of the invention to provide a method of producing a
functional Fc.sub..epsilon. RI receptor.
Further objects and advantages of the present invention will be clear from
the description that follows.
In one embodiment, the present invention relates to DNA segments that code
for polypeptides having amino acid sequences corresponding to the .alpha.,
.beta., and .gamma. subunits of Fc.sub..epsilon. RI.
In another embodiment, the present invention relates to polypeptides having
amino acid sequences corresponding to the .alpha., .beta., and .gamma.
subunits of Fc.sub..epsilon. RI.
In a further embodiment, the present invention relates to recombinant DNA
molecules comprising a vector and a DNA segment that codes for a
polypeptide having an amino acid sequence corresponding to the .alpha.,
.beta., or .gamma. subunits of Fc.sub..epsilon. RI.
In yet another embodiment, the present invention relates to a cell that
contains the above described recombinant DNA molecule.
In a further embodiment, the present invention relates to a method of
producing polypeptides having amino acid sequences corresponding to the
.alpha., .beta., and .gamma. subunits of Fc.sub..epsilon. RI.
In another embodiment, the present invention relates to a method of
producing a functional Fc.sub..epsilon. RI receptor comprising introducing
into a host cell DNA segments encoding the .alpha., .beta., and .gamma.
subunits of Fc.sub..epsilon. RI; and effecting expression of said DNA
segments under conditions such that said receptor is formed. Expression of
the receptor on the surface of COS 7 cells is achieved by the present
invention when the cDNA for all three subunits of Fc.sub..epsilon. RI are
simultaneously cotransfected. This success in expression of IgE binding
permits detailed analysis of the IgE-receptor interaction and thus enables
the development of therapeutically effective inhibitors.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1. The nucleotide sequence and predicted amino acid sequence of human
Fc.sub..epsilon. RI alpha cDNA are shown.
FIG. 2. The amino acid sequence homology of rat Fc.sub..epsilon. RI alpha
subunit (R), human Fc.sub..epsilon. RI alpha subunit (A), and mouse
Fc.sub..epsilon. RI alpha subunit (M) are shown. The regions of identity
between the three are boxed. The number one position corresponds to the
site of the predicted mature N-terminus of each protein.
FIG. 3. A flow chart showing the construction of eukaryotic expression
vectors which direct the synthesis of a complete biologically active
Fc.sub..epsilon. RI alpha chain (pHAI, pHAII) or a soluble, secreted,
biologically active Fc.sub..epsilon. RI alpha chain (pHASI, pHASII) is
presented.
FIG. 4. A flow chart showing the construction of a prokaryotic expression
vector which directs the synthesis of a soluble biologically active
Fc.sub..epsilon. RI alpha chain (which consists of amino acid residues
26-204) is presented.
FIG. 5. Restriction maps for .beta. cDNAs and strategy by which they were
sequenced. The open rectangle indicates the sequence predicted to code for
the .beta. subunit; the lines indicate the 5' and 3' untranslated regions.
The upper scheme shows the 1.5 kilobase (kb) clone containing a Pst I
cleavage site. The lower scheme shows a 2.4-kb clone containing a ClaI
cleavage site. The 3' region of the latter has been truncated as indicated
by the slashes. Its untranslated portion was sequenced as completely as
the rest of the clone. Restriction sites are indicated by vertical bars:
Hf, HinfI; Hh, Hha I; Al, Alu I; Hp, Hph I; Av, Ava II; Ac, Acc I; Ec,
EcoRI; Hd, HindIII. The horizontal arrows show the direction and extent of
sequencing by the dideoxynucleotide chain-termination method.
FIG. 6. (A)(2)-(6) Nucleotide and deduced amino acid sequences of the cDNA
coding for the .beta. subunit. Beginning at the arrowhead
(.tangle-soliddn.), an alternative sequence FIG. 6B was observed in six
clones. The putative transmembrane domains are underlined. The tryptic
peptides of the .beta. subunit, from which the amino acid sequences were
determined directly, are bracked (<>). A putative poly(A) signal near the
end is underlined. FIG. 6B Continuation of the nucleotide sequence of the
deleted form of .beta. cDNA, 3' to the junction indicated in A
(.tangle-soliddn.).
FIG. 7. Expression of cDNA coding for the .beta. subunit. (A) Comparison of
in vivo and in vitro translation products. RBL cells were grown in
[.sup.35 S] cysteine containing medium. The detergent extract of the cells
was precipitated with mAb.beta.(JRK) and, after vigorous washing,
extracted with sample buffer and electrophoresed (lane 1). This experiment
employed concentrations of detergent high enough to dissociate the
receptor completely. A transcript from the .beta. cDNA was treated in
vitro in [.sup.35 S ]methionine-containing medium (lanes 2, 3, and 5). A
control incubation contained no cDNA (lane 4). The mixtures were allowed
to react with monoclonal antibodies to the .beta. subunit after a clearing
immunoprecipitation. The specific washed precipitates were dissolved in
sample buffer and electrophoresed: lanes 2 and 4, mAb.beta.(JRK); lane 3,
mAb.beta.(NB); lane 5, irrelevant monoclonal antibody [mAb(LB)]. An
autoradiograph of the 12.5% polyacrylamide gel on which the specimens were
analyzed under reducing conditions is shown. (B) Localization of one
epitope to the NH.sub.2 -terminal peptide of the .beta. subunit. A .beta.
cDNA-containing vector was digested with HhaI before transcription using
T7 polymerase. The resulting mRNA was translated to generate an NH.sub.2
-terminal peptide of the .beta. subunit (amino acid 1-21) labeled with
[.sup.35 S]methionine. The mixture was allowed to react with
mAb.beta.(JRK) (lane 1) and the irrelevant mAb(LB) (lane 2). The
precipitates were analyzed on a 17% gel under nonreducing conditions. (C)
Expression by E. coli of a COOH-terminal fragment of the .beta. subunit. A
HinfI fragment, containing nucleotides 499-787, was subcloned into an E.
coli expression vector (Crowl et al. (1985) Gene 38:31-38) and extracts
were prepared. The proteins were electrophoresed as in A and transferred
to nitrocellulose paper. The latter was allowed to react sequentially with
monoclonal antibody mAb.beta.(NB), developed with alkaline
phosphatase-conjugated goat anti-mouse IgG (Fc), and developed in the
usual way (Rivera et al. (1988) Mol. Immunol., in press). An enlargement
of the lower half of the immunoblot is shown. Lane 1, extract from
transformant without insert; lane 2, extract from transformant with insert
in wrong direction; lane 3, extract from transformant with insert
correctly oriented. (D) Reactivity of .beta. subunits with polyclonal
antibodies induced by E. coli-expressed HinfI fragments. Purified
IgE-receptor complexes were electrophoresed, transferred to nitrocellulose
paper, and allowed to react with antibodies and subsequently with an
appropriate alkaline phosphatase-conjugated anti-immunoglobulin antibody.
Lane 1, mAb.beta.(JRK); lane 2, mAb.beta.(NB); lane 3, immune serum to
fragment A; lane 5, immune serum to fragment B; lanes 4 and 6, preimmune
sera corresponding to the immune sera in lanes 3 and 5, respectively;
lanes 7 and 8, second antibody only. This gel was run without molecular
weight standards.
FIG. 8. Hydropathicity plot of predicted sequence for the .beta. subunit.
The procedure and hydropathicity scale recommended by Engleman et al.
(Engelman et al. (1986) Annu. Rev. Biophys. Biophys. Chem. 15:321-353) was
used. The net hydropathicity value for the 20 amino acids for each
successive "window" is plotted at the position corresponding to the 10th
residue. A net free energy of >20 kcal (1 cal=4.18 J) for transfer to
water suggests a transmembrane segment (Engelman et al. (1986) Annu. Rev.
Biophys. Biophys. Chem. 15:321-353).
FIG. 9. Nucleotide sequence of the .gamma. subunit of rat Fc.sub..epsilon.
RI and the amino acid sequence that it predicts. The putative
transmembrane domain is underlined. Amino acid resides are numbered
starting with the first residue of the mature protein. Residues 5' to
residue 1 have negative numbers and include the residues encoding a
putative signal peptide according to the criteria of G. von Heijne
(Nucleic Acids Res. 14:4683-4690 (1986)). The N-terminal and C-terminal
cleavage sites are indicated by an arrow. The four tryptic peptides which
were covered and sequenced are bracketed. An asterisk denotes an ambiguous
residue in the sequence of the first tryptic peptide.
FIG. 10. Hydropathicity plot of predicted sequences of Fc.sub..epsilon. RI:
.alpha. subunit (panel A), .beta. subunit (panel B) and .gamma. subunit
(panel C). The hydropathicity scale is according to Engelman et al (Ann.
Rev. Biophys. Biophys. Chem. 15:321-353 (1986)). The summed hydropathicity
values for the 20 amino acids in successive "windows" is plotted at the
position corresponding to the tenth residue.
FIGS. 11A-11D. Formation of IgE rosettes by transfected COS 7 cells and RBL
cells. COS 7 cells were cotransfected with the coding portions of .alpha.,
.beta. and .gamma. cDNAs and sensitized with mouse IgE anti-DNP before
being exposed to red cells derivatized with TNP (FIG. 11A). As a positive
control, RBL cells were similarly tested for rosette formation (FIG. 11B).
The specificity of the rosetting assay was assessed by preincubating the
cotransfected COS 7 cells (FIG. 11B) and RBL cells (FIG. 11D) with rat IgE
(which lacks the anti-DNP activity) prior to the addition of the mouse
anti-DNP IgE.
FIG. 12. Model of the tetrameric high affinity receptor for IgE. The
polypeptides are shown in their fully processed form. The receptor is
oriented such that the large extracellular portion of the .alpha. subunit
is shown at the top and the remainder of the chain on the left. To the
right of the .alpha. subunit is the .beta. subunit with its four
transmembrane segments and to the right of it, the dimer of .gamma.
chains. Cysteines 26 and 68 and cysteines 107 and 151 in the .alpha. chain
are paired as they are likely to be disulfide linked, as are the
homologous cysteines in the Fc.sub..gamma. receptors (M. Hibbs et al, J.
Immunol. 140:544-550 (1988)). The putative transmembrane segments have all
been shown as consisting of 21 residues and would be expected to be in an
.alpha.-helical conformation. The single letter code for amino acids is
used (M. Dayhoff et al, in Atlas of Protein Sequence and Structure, Suppl.
3, ed. M. Dayhoff, 363-373, Natl. Biomed. Res. Fndtn., Washington D.C.
(1978)). Every 10th residue (starting from the N-terminus) is shaded.
DETAILED DESCRIPTION OF THE INVENTION
The present invention relates, in part, to DNA sequences which code for
polypeptides corresponding to the subunits of human Fc.sub..epsilon. RI.
More specifically, the present invention relates to DNA segments (for
example, cDNA molecules) coding for polypeptides having amino acid
sequences corresponding to the .alpha., .beta., and .gamma. subunits of
Fc.sub..epsilon. RI. In one embodiment, the DNA segments have the sequence
shown in FIGS. 1, 6, or 9, or allelic or species variation thereof, or a
unique portion of such a sequence (unique portion being defined herein as
at least 15-18 bases). In another embodiment, the DNA segments encode the
amino acid sequence shown in FIGS. 1, 6, or 9, or allelic or species
variation thereof, or a unique portion of such a sequence (unique portion
being defined herein as at least 5-6 amino acids).
In another embodiment, the present invention relates to polypeptides having
amino acid sequences corresponding to the .alpha., .beta., and .gamma.
subunits of Fc.sub..epsilon. RI. In one preferred embodiment, the
polypeptides have amino acid sequences as shown in FIGS. 1, 6, and 9, or
allelic or species variations thereof, or a unique portion of such
sequences (unique portion being defined herein as at least 5-6 amino
acids).
In another embodiment, the present invention relates to a recombinant DNA
molecule comprising a vector (for example--plasmid or viral vector) and a
DNA segment coding for a polypeptide corresponding to the .alpha., .beta.,
or .gamma. subunit of Fc.sub..epsilon. RI, as described above. In a
preferred embodiment, the encoding segment is present in the vector
operably linked to a promoter.
In further embodiment, the present invention relates to a cell containing
the above described recombinant DNA molecule. Suitable host cells include
procaryotes (such as bacteria, including E. coli) and both lower
eucaryotes (for example yeast) and higher eucaryotes (for example,
mammalian cells). Introduction of the recombinant molecule into the host
cell can be affected using methods known in the art.
In another embodiment, the present invention relates to a method of
producing the above described polypeptides, comprising culturing the above
described host cells under conditions such that said polypeptide is
produced, and isolating said polypeptide.
In a further embodiment, the present invention relates to a method of
producing a functional Fc.sub..epsilon. RI receptor comprising introducing
into a host cell DNA segments encoding the .alpha., .beta., and .gamma.
subunits of Fc.sub..epsilon. RI; and effecting expression of said segments
under conditions such that said receptor is formed.
The DNA sequences and polypeptides according to this invention exhibit a
number of utilities including but not limited to:
1. Utilizing the polypeptide or a fragment thereof as an antagonist to
prevent allergic response, or as a reagent in a drug screening assay.
2. Utilizing the polypeptide as a therapeutic.
3. Utilizing the polypeptide for monitoring IgE levels in patients.
4. Utilizing the DNA sequence to synthesize polypeptides which will be used
for the above purposes.
5. Utilizing the DNA sequences to synthesize cDNA sequences to construct
DNA useful in diagnostic assays.
The present invention will be illustrated in further detail in the
following examples. These examples are included for illustrative purposes
and should not be considered to limit the present invention.
EXAMPLE 1
Isolation of cDNA Clones for the Alpha Subunit of Human Fc.sub..epsilon. RI
RNA was extracted from KUB12 cells as described by Kishi, Leukemia
Research, 9,381 (1985) by the guanidium isothiocyanate procedure of
Chirgwin, et al., Biochemistry, 18,5294 (1979) and poly(A) mRNA was
isolated by oligo-dt chromatography according to the methods of Aviv, et
al., P.N.A.S. U.S.A., 69,1408 (1972). cDNA synthesis was performed as
previously described Kinet, et al., Biochemistry, 26,2569 (1987). The
resulting cDNA molecules were ligated to EcoRI linkers, digested with the
restriction enzyme EcoRI, size fractioned and ligated to .lambda.gtll
EcoRI arms as set forth in Young et al., Science, 222,778 (1983). The cDNA
insert containing .lambda.gtll DNA as packaged into bacteriophage lambda
particles and amplified on Y1090. A total of 1.2.times.10.sup.6
independent cDNA clones were obtained. The cDNA library was plated onto
Y1090 on 150 mm.sup.2 plates (10.sup.5 per plate) and transferred to
nitrocellular filters. The cDNA library filters were screened by in situ
hybridization using a nick translated cDNA fragment as in Kochan, et al.,
Cell, 44,689 (1986). The cDNA fragment was obtained from the rat
Fc.sub..epsilon. RI alpha cDNA corresponding to nucleotides 119-781.
Positive plaques were identified, purified and the cDNA inserts were
subcloned, using standard techniques, into the PGEM vectors (Promega
Biotech, Madison, Wis.). The cDNA insert was mapped by restriction enzyme
analysis, subcloned into derivatives of pGEM and sequenced using the
dideoxynucleotide method of Sanger et al., P.N.A.S., 74,5463 (1977)
following the GemSeq double strand DNA sequencing system protocol from
Promega Biotech (Madison, Wis.). The DNA sequence was determined for both
strands of the cDNA clone pLJ663 (nucleotides 1-1151) and for 300 bp of
each end of clone pLJ 587 (nucleotides 658-1198). No discrepancy in DNA
sequence between the two cDNA clones was observed.
The sequence for the human Fc.sub..epsilon. RI alpha cDNA is presented in
FIG. 1. The predicted amino acid sequence for the human Fc.sub..epsilon.
RI alpha polypeptide is shown below the nucleotide sequence, beginning
with methionine at nucleotide 107-109 and ending with asparagine at
nucleotide 875-877. The site of the predicted mature N-terminus was
determined to be valine at nucleotide 182-184 according to the rules set
forth by von Heijne, Eur. Journal of Biochem; 137,17; and Nucleic Acid
Research, 14,4683 (1986). This predicts a 25 amino acid signal peptide.
The rest of the cDNA sequence suggests that the human Fc.sub..epsilon. RI
alpha chain contains a 179-204) with 2 homologous domains (14 out of 25
residues are identical; residues 80-104 and 163-190), a 20-residue
transmembrane segment (residues 205-224) and a 33 residue cytoplasmic
domain containing 8 basic amino acids. Overall, there is 49% identity
between the human and rat Fc.sub..epsilon. RI alpha sequences, and 37%
identity between the human Fc.sub..epsilon. RI alpha and mouse FcGR alpha
(FIG. 2). The greatest level of homology is within the transmembrane
region where 9 amino acids surrounding the common aspartic acid residue
are identical.
EXAMPLE 2
Expression of the Human Fc.epsilon.RI Alpha Complete and Soluble Forms in
Eukaryotic Cells
Using the recombinant cDNA clone for the human Fc.sub..epsilon. RI alpha
chain, it is possible to introduce these coding sequences into an
appropriate eukaryotic expression vector to direct the synthesis of large
amounts of both a complete and soluble form of the alpha chain. For
surface expression it may necessary that the alpha subunit be complexed
with the beta or gamma subunit whereas for the eukaryotic expression of
the secreted form of the alpha subunit this may not be necessary. An
appropriate vector for the purpose is pBC12BI which has previously been
described in Cullen, (1987) Methods in Enzymology 152, Academic Press,
684. Construction of expression vectors coding for the complete alpha
chain can be isolated as follows (FIG. 3): A unique BqlII-SspI fragment
(nucleotides 65-898) is isolated from pLJ663, the BglII end is filled in
with DNA polymerase I Klenow fragment and ligated into pBC12BI which has
been restricted with either HindIII-BamHI or HindIII-SmaI (the ends are
made blunt by filling in with DNA polymerase I Klenow fragment). The
reason for attempting two different constructions is that the former
contains a 3' intron while the latter does not. The presence or absence of
introns may affect the levels of the alpha protein which are synthesized
in cells transfected by these vectors. Construction of expression vectors
coding for the soluble form of the alpha chain would be accomplished by
introducing a termination codon at nucleotides 719-721 of the coding
region in the alpha chain of the expression vectors noted above (pHAI,
pHAII, FIG. 3). This would remove the putative transmembrane and
cytoplasmic regions resulting in the synthesis of a secreted soluble form
of the human alpha chain. Introduction of a termination codon is
accomplished by oligonucleotide-directed site specific mutagenesis as
outlined by Morinaga et al., Bio. Tech., 2, 636 (1984). The sequence of
the oligonucleotide will be 5' AAGTACTGGCTATGATTTTTTATCCCATTG 3'. The
resulting expression vectors are pHASI and pHASII (FIG. 3) and these will
direct the synthesis of a truncated alpha protein corresponding to amino
acids 1-204. Expression of this protein in eukaryotic cells will result in
synthesis of a mature, IgE binding protein encompassing amino acid
residues 26-204.
The expression vectors are then introduced into suitable eukaryotic cells
such as CHO or COS by standard techniques such as those set forth in
Cullen, (1987), Methods in Enzymology, Academic Press, NY 152:684, in the
presence of a selectable marker such as G418 or Methotrexate resistance.
The selectable marker for Methotrexate resistance has an added advantage,
since the levels of expression can be amplified by introducing the cells
to higher levels of drugs. The synthesis of protein is monitored by
demonstrating the ability of human IgE (or rat IgE) to bind to these cells
(in the case of the complete alpha chain), or in the case of the soluble
form of the alpha chain, to demonstrate that the protein secreted from
these cells has the ability to bind IgE in the presence or absence of the
beta.
EXAMPLE 3
Expression of the Human Fc.sub..epsilon. RI Alpha Soluble Form in
Prokarvotic Cells
Using the recombinant cDNA clone for the human Fc.sub..epsilon. RI alpha
chain, it is possible to introduce these coding sequences into an
appropriate prokaryotic expression vector to direct the synthesis of large
amounts of a soluble (non-membrane bound) IgE binding polypeptide derived
from the alpha chain. An appropriate vector for this purpose is pEV-1
which has been described by Crowl, et al., Gene, 38, 31 (1985).
Construction of an expression vector coding for a soluble alpha chain can
be isolated as set forth in FIG. 4: a unique MstII-SspI fragment
(nucleotides 195-898 is isolated from pLJ663, the MstII end is filled in
with DNA polymerase I Klenow fragment and ligated into pEV-1 which has
been restricted with EcoRI, and the ends filled in with Klenow (FIG. 4,
pEVA). The N-terminus of the mature alpha chain is reconstructed by
oligonucleotide directed-site specific mutagenesis. The sequence of the
oligonucleotide will be 5' GAATTAATATGGTCCCTCAGAAACCTAAGGTCTCCTTG 3'.
Introduction of this sequence into the expression vector pEVA aligns the
Methionine residue of the EV-1 vector next to Valine-26 (the predicted
mature N-terminus of the alpha chain) followed by amino acid residues
27-204 (pEVHA, FIG. 4). Reconstruction of the soluble form
Fc.sub..epsilon. RI alpha is accomplished by oligonucleotide site-directed
mutagenesis. The sequence of the oligonucleotide will be
5'-AAGTACTGGCTATGATTTTTTATCCCATTG-3'. Introduction of this sequence into
the expression vector, terminates polypeptide synthesis just prior to the
start of the transmembrane region. The protein thus encoded by expression
vector pEVHAS, should faithfully direct the synthesis of a soluble form of
the alpha chain, corresponding to amino acid residues 26-204. This
expression vector is then transformed into suitable hosts.
EXAMPLE 4
Isolation and Sequence Analysis of Peptides of the Beta Subunit of
Fc.sub..epsilon. RI
Since repeated attempts to sequence intact .beta. chains were unsuccessful,
peptides were isolated from tryptic digests. Electroeluted .beta. subunits
from polyacrylamide gels were prepared as described (Alcaraz et al. (1987)
Biochemistry 26:2569-2575). Tryptic peptides were separated by
high-pressure liquid chromatography and sequenced as before (Kinet et al.
(1987) Biochemistry 26:4605-4610). A peptide (no. 1) isolated from an
initial digest had the sequence
Tyr-Glu-Glu-Leu-His-Val-Tyr-Ser-Pro-lle-Tyr-Ser-Ala-Leu-Glu-Asp-Thr. The
same peptide from later digests showed an additional leucine at the
NH.sub.2 terminus and an arginine at the COOH terminus. The sequences of
three other peptides, each isolated in substantial yields, are indicated
in a subsequent figure.
EXAMPLE 5
Cloning and Sequencing of cDNA clones of the Beta Subunit of
Fc.sub..epsilon. RI
RNA extracted from rat basophilic leukemia (RBL) cells by the guanidinium
isothiocyanate method (Chirgwin et al. (1979) Biochemistry 18:5294-5299)
was fractionated on an oligo(dT)-cellulose column (Maniatis et al. (1982)
Molecular Cloning: A Laboratory Manual (Cold Spring Harbor Lab., Cold
Spring Harbor, N.Y.) and used to construct a pUC-9 and a .lambda.gtll
library (Maniatis et al. (1982) Molecular Cloning: A Laboratory Manual
(Cold Spring Harbor Lab., Cold Spring Harbor, N.Y.; Young and Davies
(1983) Proc. Natl. Acad. Sci. USA 80:1194-1198). The initial sequence
obtained for peptide 1 was used to construct two 26-mer oligonucleotides
of 32-fold degeneracy:
5'-GGIGA(A/G)TA(G/C)ACATGIA(A/G)(C/T)TC(C/T)TCATA-3' and
5'-GGICT(A/G)TA(G/C)ACATGIA(A/G)(C/T)TC(C/T)TCATA-3'. A .lambda.gtll
library constructed from mRNA of RBL cells was screened with 1:1 mixture
of these oligonucleotides. Colonies were screened as before (Kinet et al.
(1987) Biochemistry 26:4605-4610) using oligonucleotides prepared on a
model 380A automated DNA synthesizer (Applied Biosystems, Foster City,
Calif.). Six positive clones gave similar restriction patterns. cDNA
inserts were subcloned into pGEM-4 or pGEM-3Z and the resulting
double-stranded DNA was sequenced with the Gemseq/RT sequencing system
according to the method recommended by the supplier (Promega Biotec,
Madison, Wis.). Twenty-mer oligonucleotides, corresponding to previously
sequenced regions by this method, were used as primers to generate
overlapping sequences otherwise difficult to obtain. In some instances,
DNA sequencing was performed using Sequenase as recommended by the
supplier (United States Biochemical, Cleveland). The clone containing the
longest insert was sequenced according to the strategy shown in the upper
portion of FIG. 5. The sequence predicts possible starting codons at
nucleotides 46-48 and 55-57, which would yield a polypeptide of 246 or 243
residues, respectively (FIG. 6A). The predicted M.sub.r of about 27,000 is
some 20% less than the apparent molecular weight of .beta. subunits when
analyzed on polyacrylamide gels (Holowka and Metzger (1982) Mol. Immunol.
19:219-227). In addition, no in-frame stop codon was apparent upstream of
the start codon. To rule out the possibility that the true start codon was
still further 5', the cDNA library was rescreened with a restriction
fragment (nucleotides 7-474) and with a synthetic oligonucleotide probe
(nucleotides 3-32). Twenty-eight additional clones were isolated and their
restriction patterns were examined. Twenty were similar to the original
clones. Only six additional nucleotides at the 5' end (nucleotides 1-6.
FIG. 6A) were identified. Early termination was found in six clones, which
otherwise had the same sequence through nucleotide 375 (FIG. 6B). One
2.4-kb clone had cytidine 473 substituted with an adenine. This
substitution abolishes the Pst I site and creates a new Cla I site at
nucleotide 470. Also thereby, Ala-140 would become Asp-140 (FIG. 6A).
Finally, one clone extended .apprxeq.350 base pairs (bp) in the 5'
direction. The junction with the sequence shown in FIG. 6A was
AATAAAACAAAAAAAAAAAAATG, the last two nucleotides of the newly generated
ATG corresponding to nucleotides 8 and 9 of the previous sequence. It is
likely that this clone simply resulted from the ligation of two
independent cDNAs. Screening of the pUC-9 library revealed three clones.
However, the sequence of none of these extended 5' beyond nucleotide 84.
EXAMPLE 6
RNA Transfer Blotting
RNA transfer blotting was performed under high stringency using a Pst I
fragment probe (nucleotides 1-474). Thirty micrograms of total RNA was run
on a 1% agarose gel containing 2% formaldehyde and blotted to
nitrocellulose filters (Maniatis et al. (1982) Molecular Cloning: A
Laboratory Manual (Cold Spring Harbor Lab., Cold Spring Harbor, N.Y.). The
filters were hybridized with a restriction fragment of the .beta. cDNA
(nucleotides 1-474) as described (Maniatis et al. (1982) Molecular
Cloning: A Laboratory Manual (Cold Spring Harbor Lab., Cold Spring Harbor,
N.Y.) and washed with 15 mM NaCl/1.5 mM sodium citrate at 65.degree. C.
RBL cells yielded two major bands at .apprxeq.2.7 kb and 1.75 kb with the
upper band having about twice the intensity of the lower one. A minor band
1.2 kb was also noted. Negative results were obtained with a variety of
cells that do not express high-affinity IgE receptors: the rat pituitary
line GH3 (American Type Culture Collection no. CCL82.1), the rat glial
cell line C6 (no. CCL107), the mouse Leydig cell line I-10 (no. CCL83),
and, notably, the mouse monocytic line J774 (no. T1B67) and the rat
lymphoma "NTD" (Rivera et al. (1988) Mol. Immunol., in press).
EXAMPLE 7
In vitro Transcription and Translation
cDNAs corresponding to the .beta. subunit and various mutated or truncated
forms thereof were subcloned into either pGEM-4 or pGEM-3Z transcription
vectors (Promega Biotec). The .beta. clone containing the Pst I site was
transcribed in vitro with T7 RNA polymerase. Unlabeled RNAs were
synthesized using either SP6 or T7 polymerase as recommended by the
supplier. Capping reactions were performed as reported (Contreras et al.
(1982) Nucleic Acids Res. 10:6353-6362). After digestion of the template
with RNase-free DNase I, the RNAs were purified further by extraction with
phenol/chloroform and three precipitations from ethanol. The RNA was then
translated with a micrococcal nuclease-treated lysate of rabbit
reticulocytes in the presence of [.sup.35 S] methionine as recommended by
the supplier (Promega Biotec). The products of translation were diluted
1:1 with 20 mM detergent {3-[3-(cholamidopropyl)dimethylammonio]-1-propane
sulfonate in borate-buffered saline (pH 8) containing 30 .mu.l of
aprotinin per ml, 175 .mu.g of phenylmethyl-sulfonyl fluoride per ml, 10
.mu.g of leupeptin per ml, and 5 .mu.g of pepstatin per ml and
immunoprecipitated with monoclonal antibodies as described (Rivera et al.
(1988) Mol. Immunol., in press). The unfractionated translated material
showed a major component at M.sub.r .apprxeq.32,000 compared to the
control from which the RNA had been omitted or an alternative RNA (brome
mosaic virus) had been substituted (data not shown).
The isolation of antibodies was as follows: Escherichia coli transformed
with an expression vector containing the desired restriction fragments
(Crowl et al. (1985) Gene 38:31-38; Portnoy et al. (1986) J. Biol. Chem.
261:14697-14703) were cultured and induced, and the fraction enriched for
the recombinant protein was prepared as described (Portnoy et al. (1986)
J. Biol. Chem. 261:14697-14703). After separation on polyacrylamide gels
in sodium dodecyl sulfate (NaDodSO.sub.4) the transformant-specific
protein was eluted and used to immunize rabbits. Approximately 100 .mu.g
of protein was injected in complete Freund's adjuvant; this was followed
by a booster injection of 25 .mu.g of protein in incomplete adjuvant. The
isolation and characterization of monoclonal anti-.beta. antibodies
mAb.beta.(JRK) and mAb.beta.(NB) (the latter, a generous gift from David
Halowka, Cornell University) have been described (Rivera et al. (1988)
Mol. Immunol., in press).
The monoclonal anti-.beta. antibodies mAb.beta.(JRK) and mAb.beta.(NB)
(Rivera et al. (1988) Mol. Immunol., in press) (FIG. 7A, lanes 2 and
3)--but not an irrelevant antibody (lane 5)-precipitated radioactive
material, which on polyacrylamide gels in NaDodSO.sub.4 showed a major
band at M.sub.r 32000. This band had the identical mobility as the upper
band of the doublet precipitated by mAb.beta.(JRK) from an extract of
labeled RBL cells (lane 1). Although not seen well in the reproduction,
the autoradiogram showed that the material synthesized in vitro also
contained the lower molecular weight component seen the in vivo
synthesized .beta. chains. The mobility of the in vitro synthesized
protein was unaltered by reduction as has been previously observed with
the .beta. subunit. The clone containing the Cla I site (which lacks the
first ATG codon) led to the synthesis of a protein whose mobility on gels
was indistinguishable from that for the clone containing the Pst I site.
On the other hand, an aberrant clone containing the newly generated ATG
(above) induced the synthesis of a somewhat larger protein with an
apparent M.sub.r of 33,500 (data not shown). In vitro translation of a
transcript coding for the NH.sub.2 -terminal 21 amino acids of the .beta.
subunit led to a product precipitable by mAb.beta.(JRK) (FIG. 7B).
EXAMPLE 8
Expression of the Beta Subunit of Fc.sub..epsilon. RI in E. Coli
Two HinfI fragments (A, nucleotides 106-498; B, nucleotides 499-787) were
individually subcloned into an E. coli expression vector, and extracts
were prepared from the induced cultures. The results of one immunoblotting
experiment are shown in FIG. 7C. The material extracted from the bacteria
transformed with a vector containing the HinfI fragment B exhibited a
M.sub.r 14,000 component reactive with mAb.beta.(NB) but not with
mAb.beta.(JRK) (FIG. 7C, lane 3). The extract from the transformants
containing the more NH.sub.2 -terminal HinfI fragment A (residues 17-148)
reacted with neither antibody (compare with above). Rabbit antibodies
generated by fragment A reacted on immunoblots with purified receptors
exactly at the position where the two monoclonal anti-.beta. antibodies
reacted (FIG. 7D, lanes 1-3) and quantitatively precipitated intact
.sup.125 I-labeled IgE-receptor complex from unfractionated detergent
extracts of RBL cells (data not shown).
EXAMPLE 9
Biosynthetic Incorporation
Biosynthetic incorporation of labeled amino acids and monosaccharides was
as described (Perez-Montfort et al. (1983) Biochemistry 27:5722-5728). The
purification and analysis on gels and by immunoblotting of the
IgE-receptor complexes have also been described (Rivera et al. (1988) Mol.
Immunol., in press).
By using biosynthetic incorporation of two different amino acids labeled
distinguishably, their ratio in the subunits of the receptor (Table 1,
right part) was determined. The ratios of four distinctive amino acids to
each other was in satisfactory agreement with the ratios predicted from
the .beta. cDNA clone (Table 1, right part, columns 1-3). Because the cDNA
for the .beta. subunit predicts three potential glycosylation sites, a
double-labelling experiment using [.sup.3 H]mannose and [.sup.35
S]cysteine was also performed. Based on the relative carbohydrate data
reported for the .alpha. subunit (Kaneilopoulos et al. (1980) J. Biol.
Chem. 255:9060-9066) and correcting them on the basis of the peptide
molecular weight for this chain predicted from the cDNA, it was calculated
that the .alpha. subunit contains .apprxeq.20 mol of mannose per mol. It
was therefore possible to determine the mannose/cysteine ratio in the
.beta. subunit from the double-labeling experiment. The results showed
only 0.05 mol/mol of cysteine or 0.3 mol/mol of the .beta. subunit (Table
1, right part, column 4).
TABLE 1
__________________________________________________________________________
Amino Acid compositon of .beta. Subunits
__________________________________________________________________________
cDNA versus compositional analysis for the 8 subunit
Asx
Thr
Ser
Glx
Pro
Gly
Ala
Val
Met
Ile
Leu
Tyr
Phe
His
Lys
Arg
Cys
Trp
__________________________________________________________________________
Deduced from 20 12 23 24 15 12 19 17 4 15 36 9 12 1 8 8 6 2
.beta.cDNA
Direct analysis* 22 13 22 27 13 19 18 14 4 13 31 7 10 2 10 10 5 ND
Double-labeli
ng
studies.sup..dagger.
__________________________________________________________________________
cDNA versus incorporation
data
Met/His
Cys/His
Cys/Trp
Man/Cys
__________________________________________________________________________
Deduced from 4 6 3 --
.beta.cDNA
Direct analysis*
Double-labeling 4.2 5.1 2.5 0.05
studies.sup..dagger.
__________________________________________________________________________
*The mol % of each amino acid as reported by Alcaraz et al. ((1987)
Biochemistry 26: 2569-2575) was multiplied by 241 the number of residues
excluding tryptophan predicted from the cDNA. ND, not determined.
.sup..dagger. *IgEreceptor complexes were purified from RBL cells
incubated with a mixture of two precursors labeled with differentiable
radioisotopes. The subunits were separated on a polyacrylamide gel. The
gel was sectioned into 2mm slices, extracted, and assayed for
radioactivity by scintillation spectroscopy. The ratio of cpm of .sup.35
S/.sup.3 H was individually calculated for .alpha., .beta., and .gamma.
subunits. The ratio in the .alpha. subunit is proportional to the known
molar ratio of the # .sup.35 Slabeled and .sup.3 Hlabeled residues in th
.alpha. subunit. Hence, the corresponding ratio in the .beta. subunit (an
the .gamma. subunit) predicts the ratio of the same residues in the latte
subunits.
EXAMPLE 10
Seauence Characteristics
There is ample evidence that the cDNAs that were isolated code for the
.beta. subunit. (i) In vitro transcription of the cDNA and translation of
the derived mRNA produce a protein whose apparent molecular weight on gel
electrophoresis is indistinguishable from that of authentic .beta. chains
(FIG. 7A). (ii) The cDNA accurately predicts the sequence of four peptides
isolated from a tryptic digest of .beta. chains (FIG. 6A) and a
composition that agrees well with direct analyses and biosynthetic
incorporations (Table I). (iii) Two monoclonal antibodies reactive with
discrete epitopes on the .alpha. subunit (Rivera et al. (1988) Mol.
Immunol., in press) precipitate the protein synthesized in vitro from the
cloned cDNA (FIG. 7A), and one of them reacts with a fragment of the
protein expressed in E. coli (FIG. 7C). (iv) Polyclonal antibodies raised
against a fragment of the .beta. subunit synthesized by E. coli
transformants react with .beta. chains on immunoblots (FIG. 7D) and with
IgE-receptor complex in solution.
The nucleotide sequence at the 5' end of the cloned cDNA (clone 1) does not
in itself define the start of the open reading frame unambiguously. There
is no leader sequence and no "in frame" stop codon preceding the
presumptive start codon. In addition, the molecular weight deducted from
the cDNA (M.sub.r 27,000) is substantially lower than the one observed on
NaDodSO.sub.4 gels (M.sub.r 32,000), although the .beta. subunit is not
glycosylated. Therefore, it was possible that the start codon had been
missed. Nevertheless, the aggregate data provide strong evidence that the
full coding sequence for the .beta. subunit has been recovered. (i)
Extensive attempts failed to reveal cDNAs in either of two separate
libraries with a more extended 5' sequence. (ii) The major species
generated by 5' extension studies terminated precisely at the point at
which most of our clones started. (iii) The second ATG codon at the 5' end
meets the consensus characteristics of known initiation sites (Kozak
(1987) Nucleic Acids Res. 15:8125-8148). That it is preceded by a nearby
5' ATG codon is uncommon, but not rare (Kozak (1987) Nucleic Acids Res.
15:8125-8148), and has been observed for the human .alpha. subunit
(Shimizu et al. (1988) Proc. Natl. Acad. Sci. USA 85:1907-1911; Kochan et
al. (1988) Nucleic Acids Res. 16:3584). (iv) As already noted, in vitro
translation of an mRNA transcribed from the cDNA containing only the
second ATG codon gives a polypeptide indistinguishable in length from the
authentic .beta. chains. An aberrant clone containing a start codon 48
nucleotides 5' to the presumed start codon directed the in vitro synthesis
of a polypeptide with an apparent molecular weight appropriately greater
than that of the .beta. subunit. Therefore, the correspondence in apparent
molecular weight between authentic .beta. chains and the protein
synthesized in vitro from clone 1 is meaningful. The RNA transfer blotting
data show an mRNA of .apprxeq.2.7 kb, precisely what would be anticipated
from the cDNA that was sequenced (FIG. 6), given a poly (A) tail of
.apprxeq.200 nucleotides. In the discussion that follows it is assumed
that the B chain begins with the methionine residue coded for by the
second ATG and is, therefore, 243 residues long.
Only a single clone containing the Cla I restriction site was observed
among the 37 clones analyzed. This clone likely resulted from a single
base mutation during the cloning and is unlikely to represent a normally
occurring mRNA. Conversely, six clones showing the deleted sequence (FIG.
6B) were observed and likely reflected an authentic species of mRNA. If
translated, it would code for a M.sub.r 14,000 protein with only a single
transmembrane segment.
The sequence of the .beta. subunit contains potential sites for N-linked
glycosylation at residues 5, 151, and 154. However, past and new
incorporation data give no evidence for carbohydrate in the .beta. subunit
(Perez-Montfort et al. (1983) Biochemistry 27:5722-5728; Holowka and
Metzger (1982) Mol. Immunol. 19:219-227; and Table I). The sequence shows
no unusual features or homology to previously reported sequences, in
particular to those associated with Fc receptors or with Fc binding
factors.
A hydropathicity analysis suggests that the .beta. subunit crosses the
plasma membrane four times (FIG. 8). The hydrophilic NH.sub.2 and COOH
terminus would therefore be on the same side of the membrane. Expression
of fragments of the .beta. cDNA indicate that mAb.beta.-(NB) reacts within
amino acids residues 149-243 (FIG. 7C) and that mAb.beta.(JRK) reacts with
fragment containing residues 1-21 (FIG. 7B). Since neither antibody reacts
appreciably with intact cells but both react strongly with cell sonicates,
the combined results are consistent with the NH.sub.2 and COOH terminus
being on the cytoplasmic side of the plasma membrane.
Earlier studies had suggested that the .beta. chain contained a M.sub.r
20,000 ".beta..sub.1 " domain resistant to proteolysis while membrane
bound (Holowka and Metzger (1982) Mol. Immunol. 19:219-227). This portion
also contained those residues that were modified by an intrabilayer
labeling reagent (Holowka and Metzger (1982) Mol. Immunol. 19:219-227;
Holowka et al. (1981) Nature (London) 289:806-808) and became linked to
the .alpha. and/or .gamma. subunit when chemical crosslinking reagents
were used (Holowka and Metzger (1982) Mol. Immunol. 19:219-227) and to the
.gamma. subunit when spontaneous disulfide linkage between the .beta. and
.gamma..sub.2 subunits occurred (Kinet et al. (1983) Biochemistry
22:5729-5732). The remainder, ".beta..sub.2 ", appeared to contain the
serine residues that became phosphorylated in situ (Perez-Montfort et al
(1983) Biochemistry 22:5733-5737; Quarto and Metzger (1986) Mol. Immunol.
23:1215-1223) but has never been positively identified as a discrete
fragment. The sequence predicted by the cDNA for the .beta. subunit
suggests that part or all of either the NH.sub.2 -terminal 59 residues or
the COOH-terminal 44 residues, or of both, is cleaved off to generate the
.beta.1 fragment.
EXAMPLE 11
Contransfection Experiments
The full-length coding sequences of the .alpha. and the .beta. subunits
were cotransfected in COS 7 cells by using a vector for transient
expression. No IgE-binding sites were expressed at the surface of
transfected cells.
Studies of the receptor with low affinity for IgE on macrophages revealed a
component that could be chemically crosslinked to the IgE-binding portion
and that had an apparent molecular weight similar to the .beta. subunit of
the high-affinity receptor (Finoloom and Metzger (1983) J. Immunol.
130:1489-1491). The peptides generated from this component by protease
digestion appeared to differ from those released from .beta. subunits, but
it raised the possibility that other Fc receptors also contained
.beta.-like subunits that had heretofore escaped detection (Rivera et al.
(1988) Mol. Immunol., in press). So far, we have no evidence for this from
RNA transfer blot experiments conducted at high stringency. In particular,
J774 cells are known to contain Fc.sub..gamma. receptors whose
immunoglobulin-binding chain shows considerable homology to the .alpha.
chain of the high-affinity receptor for IgE (Kinet et al. (1987)
Biochemistry 26:4605-4610). However, it was not possible to detect mRNA
for .beta. chains by the methods that were employed. Similarly, NTD
lymphoma cells gave negative results even though they have Fc.sub..gamma.
receptors and show a low molecular weight component that reacts with
mAb.beta.(JRK) on immunoblots (Rivera et al. (1988) Mol. Immunol., in
press). It cannot be excluded that Fc.sub..gamma. receptors have
.beta.-like subunits.
EXAMPLE 12
Isolation and Sequence Analysis of Peptides of the Gamma Subunit of
Fc.sub..epsilon. RI
Fc.sub..epsilon. RI was purified by affinity chromatography using
TNP-lysine beads as described in G. Alcaraz et al, Biochemistry
26:2569-2575 (1987). The eluate was applied to sepharose 4B beads coupled
by cyanogen bromide to monoclonal anti-.beta. (JRK) (J. Rivera et al, Mol.
Immunol. 25:647-661 (1988)). After washing the beads with 2 mM CHAPS in
borate buffered saline at pH8, the bound material was eluted at 65.degree.
C. with 0.1% sodium dodecyl sulfate, phosphate buffered saline, pH 6.5.
The subunits from Fc.sub..epsilon. RI were then separated by HPLC size
chromatography, the .beta. and .gamma. containing fractions recovered,
reduced, alkylated and digested with trypsin (J.-P. Kinet et al,
Biochemistry 26:4605-4610 (1987)). The resulting peptides were separated
by HPLC reverse phase chromatography as in J.-P. Kinet et al, Biochemistry
26:4605-4610 (1987). The chromatograms from the .beta. and .gamma. digests
were compared and the non-overlapping .gamma. peptides were sequenced
(J.-P. Kinet et al, Biochemistry 26:4605-4610 (1987)).
EXAMPLE 13
Cloning and Sequencing of cDNA Clones of the Gamma Subunit of
Fc.sub..epsilon. RI
Oligonucleotide probes were synthesized according to the sequences of
peptide 3 (residues 41 to 47) and of peptide 4 (residues 54 to 62). The
sequences were GA(A/G)AA(A/G)TCIGA(T/C)GCTCTCTA and
AA(T/C)CA(A/G)GA(A/G)ACITA(T/C)GA(A/G)ACI(T/C)TIAA. The methods used to
screen the .lambda.gt11 library, to purify, subclone and sequence the
positive clones are known in the art (J.-P. Kinet et al, Biochemistry
26:4605-4610 (1987)). Peptide 3 and peptide 4 were also synthesized using
a peptide synthesizer ABI 431A. The purity of the synthetic peptides was
assessed by HPLC reverse phase chromatography, amino acid composition and
mass spectroscopy. The peptides were conjugated either to ovalbumin using
m-Maleimidobenzoyl-N-hydroxysuccinimide ester (F.-T. Liu et al,
Biochemistry 18:690-697 (1979)) at a molar ratio of 5:1 or to sepharose 4B
with cyanogen bromide. Rabbits were immunized with the
ovalbumin-conjugated peptides, the antisera collected and the antipeptide
antibodies purified by affinity chromatography using sepharose 4B
conjugated peptides. The antipeptide antibodies were tested for reactivity
with the .gamma. subunit of Fc.sub..epsilon. RI by Western blotting and
for their ability to immunoprecipitate .sup.125 I-IgE receptor complexes
(J. Rivera et al, Mol. Immunol. 25:647-661 (1988)).
The nucleotide sequence of the .gamma. subunit of rat Fc.sub..epsilon. RI
obtained using the method of this invention, as well as the amino acid
sequence that it predicts, are shown in FIG. 9.
In order to isolate and characterize the cDNA for the .gamma. subunit,
cDNAs for the Fc.sub..epsilon. RI .gamma. subunit were isolated from a
.lambda.gt11 library prepared from rat basophilic leukemia (RBL) cells
(J.-P. Kinet et al, Biochemistry 26:4605-4610 (1987)) using
oligonucleotide probes. Four peptide sequences were identified in a
tryptic digest of the Fc.sub..epsilon. RI .gamma. subunits, and two of the
peptides were used to synthesize two oligonucleotide probes (FIG. 9). The
library was screened in duplicate with these two probes and overlapping
plaques identified. Three discrete plaques were purified, subcloned and
found to contain similar inserts of 0.6 to 0.7 kilobases (kb).
FIG. 9 shows the complete nucleotide sequence of the .gamma. cDNA, the
deduced amino acid sequence and the position in the sequence of the four
original tryptic peptides. Analysis of the sequence (FIG. 10C) indicates
an N-terminal hydrophobic signal peptide of 18 residues and a putative
transmembrane domain separating a short extracellular portion of 5
residues from an intracytoplasmic domain. As predicted by earlier studies,
the N-terminal processed .gamma. subunit contains two cysteines, no
methionine and no tryptophan residues (G. Alcaraz et al, Biochemistry
26:2569-2575 (1987)). Compositional analysis suggested that the .gamma.
subunit might contain one histidine residue (G. Alcaraz et al,
Biochemistry 26:2569-2575 (1987)). However, recent biosynthetic dual
labeling studies of the receptor using .sup.35 S methionine and .sup.3 H
histidine, clearly indicated that no trace of histidine was incorporated
into the receptor-associated .gamma. subunit. Since the open reading frame
derived from three independent clones, each predicts a histidine six
residues from the C-terminal end, it is expected that the .gamma. subunit
undergoes a C-terminal processing which clips off the histidine-containing
segment. Furthermore, because the peptide immediately preceding this
histidine was recovered (FIG. 9), the C-terminal segment must be cleaved
after Lys 63. The predicted molecular weight of the fully processed
.gamma. would therefore be 7139 Da, in close agreement with values
obtained for the purified reduced .gamma. on sodium dodecyl sulfate--urea
gels (G. Alcaraz et al, Biochemistry 26:2569-2575 (1987)).
Polyclonal antipeptide antibodies to a heptamer and to a nonamer peptide of
the .gamma. subunit (FIG. 9) were prepared and tested for reactivity with
IgE-receptor complexes for RBL cells. Both purified antipeptide antibodies
reacted in a Western blot assay with the unreduced dimer and the reduced
monomer of partially purified .gamma. subunits. In addition, both
antibodies quantitatively precipitated receptor-bound .sup.125 I-IgE,
either from an extract of RBL cells or from a preparation of partially
purified receptors. Taken together, these results leave no doubt that the
cDNAs isolated according to the present invention code for the .gamma.
subunit of Fc.sub..epsilon. RI.
EXAMPLE 14
Expression of Receptor
In order to achieve expression of the receptor on the surface of COS 7
cells, the coding region of the .alpha., .beta., and .gamma. cDNAs were
first subcloned separately into the SV 40 promoter-driven expression
vector pSVL, prior to transfection into the COS-7 cells. The 810 bp
EcoRI-Sty I restriction fragment of the .alpha. cDNA, the 965 bp
EcoRI-EcoRV restriction fragment of the .beta. cDNA and the 300 bp
EcoRI-Dde I restriction fragment of the .gamma. cDNA were subcloned
separately into the Sma I site of the transient expression vector pSVL
(Pharmacia, Uppsala, Sweden). These restriction fragments individually
contained the entire coding sequence of the appropriate subunit and
variable portions of untranslated sequences. The only foreign sequence was
the starting EcoRI recognition sequence which belonged to the initial
linker. Cultured COS 7 monkey kidney cells were then transfected with 40
.mu.g of DNA by the standard calcium phosphate precipitation technique (L.
Davis et al, in Basic Methods in Molecular Biology, ed. L. Davis,
Elsevier, N.Y. (1986)). After 48 hrs, the transfected cells (panels A and
B of FIG. 11), as well as RBL cells (panels C and D of FIG. 11), were
examined for surface expression of IgE binding by an IgE resetting assay.
The cells (5.times.10.sup.6 cells/ml) were incubated at room temperature
with (panels B and D) or without (panels A and C) 50 .mu.g/ml of
non-specific rat IgE for 30 min and then with 5 .mu.g/ml of anti-DNP-IgE
(F.-T. Liu et al, J. Immunol. 124:2728-2736 (1980)). The cells were then
rosetted with ox red blood cells that had been modified with
2,4,6-trinitrobenzene sulfonic acid according to a known method (M.
Rittenberg et al, Proc. Soc. Exp. Biol. Med. 132:575-581 (1969)). The
results are shown in FIG. 11. FIG. 11A shows IgE-binding activity
expressed by cells cotransfected with the .alpha., .beta. and .gamma.
subunits. Virtually all RBL cells, used as a positive control, formed
rosettes (FIG. 11C). The rosettes were completely inhibited by
preincubation of the cells with rat IgE (FIGS. 11B and D) but not with
human IgE (not shown). This coincides with the species specificity for the
rat Fc.sub..epsilon. RI (A. Kulczycki et al, J. Exp. Med. 139:600-616
(1974)).
In order to study the requirements for surface expression of IgE-binding
activity, the cells were transfected with different combinations of the
cDNAs for the three subunits, as shown in Table 2.
COS-7 cells were transfected with different combinations of cDNAs for the
three subunits of Fc.sub..epsilon. RI (FIG. 11). The rosetting assay was
performed for each transfection shown in Table 2. The assessment of the
mRNA by Northern blotting was performed one time only (on 2.times.10.sup.7
cells). Inhibitor was added to the cells in the experiments marked by an
asterisk in Table 2 (50 .mu.g/ml of non-specific rat IgE was added to the
cells 30 minutes prior to the addition of the specific mouse anti-DNP
IgE).
TABLE 2
______________________________________
Transfection Experiments
Expression
IgE Binding
Transfections
Receptor (rosettes/cells
Cells cDNA No. mRNA counted)
______________________________________
COS 7 0 9 0 0/12,948
.alpha. 2 .alpha. 0/4,050
.alpha..beta. 2 .alpha..beta. 0/3,504
.alpha. 4 .alpha. 0/8,030
.beta. 1 .beta. 0/2,069
.alpha..beta. 29 .alpha..beta. 920/41,238
.alpha..beta. 4 .alpha..beta. 0/7,542*
RBL 0 -- .alpha..beta. "100%"
______________________________________
*Experiments where inhibiter was added.
Table 2 summarizes the data derived from all the transfection experiments
performed by the present inventors to the time of filing the present
application. The success rate of the transfection experiments has improved
so that there is now routinely achieved 5.+-.2% expression of IgE binding
when .alpha., .beta. and .gamma. are simultaneously cotransfected.
Successful transfection was achieved for all combinations, as assessed by
Northern blotting, but rosette forming cells were only detected after
cotransfection of the full set of the cDNAs. These results indicate that
the .beta. and .gamma. subunits are required for surface-expression of the
IgE-binding .alpha. subunit. It is further indicated that only the fully
assembled receptor reaches the plasma membrane. This phenomenon has also
been observed in other systems (M. McPhaul et al, Proc. Natl. Acad. Sci.
USA 83:8863-8867 (1986); Y. Minami et al, Proc. Natl. Acad. Sci. USA
84:2688-2692 (1987)) and may be generally applicable to polymeric membrane
proteins.
The easy dissociability of .beta. and .gamma..sub.2 from .alpha. (B. Rivnay
et al, Biochemistry 21:6922-6927 (1982)) has raised persistent uncertainty
about whether conceptually, .gamma..sub.2 and .beta. should be considered
as subunits of Fc.sub..epsilon. RI or as "receptor associated" proteins.
(An example of the latter is the CD3 complex which associates with the
antigen receptor on thymus-derived lymphocytes (H. Clevers et al, Ann.
Rev. Immunol. 6:629-662 (1988)). The subunit model for Fc.sub..epsilon. RI
has been favored, for example, on the basis of the coordinate biosynthesis
and catabolism of .alpha., .beta. and .gamma..sub.2 (R. Quarto et al,
Molec. Immunol. 22:1045-1052 (1985)). The new data on transfected cells
obtained by the present invention provides the strongest evidence yet
obtained that .alpha..beta..gamma..sub.2 is the minimal structure for
Fc.sub..epsilon. RI.
The present model for the tetrameric Fc.sub..epsilon. RI receptor is
illustrated in FIG. 12. In this model each of the 589 amino acid residues
of which the expressed receptor is composed is shown as a circle. In the
diagram, the exterior of the cell would be at the top, the plasma membrane
in which the receptor is embedded would be in the middle, and the interior
of the cell towards the bottom. Each of the polypeptide chains (the
.alpha. on the left, the .beta. chain in the middle and the two .gamma.
chains on the right) contains one or more transmembrane segments.
The .alpha. chain is believed to contain two intrachain disulfide loops,
and the sequences of these loops show considerable homology with
immunoglobulins (J.-P. Kinet et al, Biochemistry 26:4605 (1987); A.
Shimizu et al, Proc. Natl. Acad. Sci. USA 85:1907 (1988); J. Kochan et al,
Nucleic Acids Res. 16:3584 (1988)). Thus, the .alpha. subunit is another
member of the immunoglobulin superfamily (A. Williams et al, Ann. Rev.
Immunol. 6:381 (1988)). The extracellular and transmembrane segments of
the .alpha. chain show considerable homology with the immunoglobulin
binding chain of Fc receptors that bind IgG (J. Ravetch et al, Science
234:178 (1986)) but the intracellular cytoplasmic tail is quite different.
The carbohydrate residues that are covalently attached to the
extracellular portion of the .alpha. chain are not indicated in FIG. 12.
There are seven potential sites for N-linked carbohydrates (J.-P. Kinet et
al, Biochemistry 26:4605 (1987); A. Shimizu et al, Proc. Natl. Acad. Sci.
USA 85:1907 (1988)), but which of these that are actually used by the cell
remains to be determined. Studies show that the carbohydrate is not
essential for the binding of IgE by this chain (B. Hempstead et al, J.
Biol. Chem. 256:10717 (1981)).
The .beta. chain contains four transmembrane segments (J.-P. Kinet et al,
Proc. Natl. Acad. Sci. USA 85:6483 (1988)) and previous studies with
monoclonal antibodies (J.-P. Kinet et al, Proc. Natl. Acad. Sci. USA
85:6483 (1988); J. Rivera et al, Mol. Immunol. 25:647 (1988)) show that
the amino- and carboxyl-termini which are respectively 59 and 43 residues
long, protrude from the cytoplasmic face of the plasma membrane.
Similarly, the .gamma. chains have an extensive intracellular extension
but only very limited exposure to the exterior.
According to the present model, the putative transmembrane domains of the
individual subunits are predicted from their respective hydropathicity
plots (see FIG. 10, wherein a net free energy of >20 kcal/mol for transfer
to water suggests a transmembrane segment or a leader peptide (D. Engelman
et al, Ann. Rev. Biophys. Biophys. Chem. 15:321-353 (1986)). These plots
suggest one, four and one hydrophobic domains for the .alpha., .beta. and
each .gamma., respectively (i.e., seven transmembrane domains for the
entire receptor). Members of a family of receptors interacting with G
proteins also contain seven transmembrane domains (I. Herskowitz et al,
Cell 50:995-996 (1987)). This family includes .beta. and .alpha.
adrenergic, muscarinic receptors and rhodopsin. Although no sequence
homology between Fc.sub..epsilon. RI and these receptors is found, it is
significant that an interaction between Fc.sub..epsilon. RI and G proteins
has been postulated to explain at least some of the biochemical pathways
activated by this receptor (S. Cockcroft et al, Nature 314:534-536
(1985)). The topology of the .alpha. and .beta. subunits has been
discussed in J.-P. Kinet et al, Biochemistry 26:4605-4610 (1987) and A.
Shimizu et al, Proc. Natl. Acad. Sci. USA 85:1907-1911 (1988), in
particular, the cytoplasmic localization of the C- and N- terminal
portions of the .beta. subunit. Two pieces of evidence support the
topology of the .gamma.-dimer as shown in FIG. 12: The .gamma. can be
oxidatively iodinated on inverted vesicles but not on intact cells (D.
Holowka et al, J. Biol. Chem. 259:3720-3728 (1984)) and, in vivo, .gamma.
becomes phosphorylated on threonine residues (R. Quarto et al, Mol.
Immunol. 23:1215-1223 (1986)). None of the relevant residues are present
in the small presumptive extracytoplasmic segment of .gamma. but all are
present on the presumptive cytoplasmic tail, i.e., two tyrosine and four
threonine residues.
As a further means to examine the topology of the receptor, the putative
extracellular and intracellular segments of the three subunits were
analyzed for their relative content of basic residues, as suggested by G.
von Heijne Biochim. Biophys. Acta 947:307-333 (1988). He found the ratio
of basic/total residues varies as a function of the length of the segment
studied, but in general was substantially higher in the non-translocated
(intracellular) segments than in the translocated (extracellular) segments
of membrane proteins. Table 3 below shows a good correspondence between
the ratios calculated for the present model and the ratios expected on the
basis of "known" membrane proteins (G. von Heijne, Biochim. Biophys. Acta
947:307-333 (1988)), thereby providing independent support for the
topological model presented here.
TABLE 3
__________________________________________________________________________
Ratio Lys + Arg/total in Translocated and Untranslocated
Segments of Receptor Subunits
Extracellular Intracellular
(translocated) (untranslocated)
No. Ratio No. Ratio
Polypeptide
residues
found
expected residues
found
expected
__________________________________________________________________________
.alpha. 179 0.13
0.11 22 0.31
0.19
.beta. Loop 1 17 0.06 0.04 N-term 59 0.10 0.10
Loop 3 28 0.03 0.04 Loop 2 12 0.25 0.20
C-term 43 0.12 0.18
.gamma. 5 0 0.08 36 0.22 0.16
.alpha..beta..gamma..sub.2 234 0.045 0.02-0.06 208 0.17 0.12-0.16
__________________________________________________________________________
The expected values calculated from the data in FIG. 8 of G. von Heijne,
Biochim. Biophys. Acta 947, 307-333 (1988), in which the ratio found for
the extramembrane segments from "known" proteins has been plotted as a
function of the segments' length.
The present model clarifies several important features with respect to the
organization of the subunits. The .beta. and dimer of .gamma. interact
with each other; in detergent solutions they dissociate from the .alpha.
as a unit before dissociating from each other (J. Rivera et al, Mol.
Immunol. 25:647-661 (1988)), and occasionally, .beta. and the .gamma.
dimer are observed to be disulfide-linked to each other (J.-P. Kinet,
Biochemistry 22:5729-5732 (1983)). The likeliest candidates for this bond
are .gamma.-cys7 and .beta.-cys80 which are predicted to be topologically
close. This would then require that at least the .gamma.-cys26 residues
are disulfide-linked in the .gamma. dimer. Preliminary data on the
receptor biosynthesis suggest that .alpha. and .beta. interact with each
other.
The functional properties of Fc.sub..epsilon. RI are broadly similar to
those of several Fc.sub..gamma. R. Fc.sub..gamma. R appears to bind to
homologous segments of the immunoglobulin's Fc region (B. Helm et al,
Nature 331:180-183 (1988); A. Duncan et al, Nature 332:563-564 (1988)),
and the binding site on the receptor is found on a homologous polypeptide
having immunoglobulin-like domains (J.-P. Kinet et al, Biochemistry
26:4605-4610 (1987); J. Ravetch et al, Science 234:718-725 (1986)). Both
types of receptors need to be aggregated to initiate cell activation and,
where studied, the latter appears to involve generation of broadly similar
second messengers (H. Metzger et al, Ann. Rev. Immunol. 4:419-470 (1986);
N. Hogg, Immunol. Today 9:185-187 (1988)). It is surprising, therefore,
that whereas Fc.sub..epsilon. RI consists of four polypeptide chains,
seven transmembrane segments and five cytoplasmic segments, Fc.sub..gamma.
Rs appear to perform similar functions with a much simpler structure,
i.e., an .alpha.-like subunit alone. The extreme case is that of
Fc.sub..gamma. RIII which appears to lack even transmembrane and
intracellular segments (P. Selvaray et al, Nature 333:565-567 (1988); D.
Simmons et al, Nature 333:568-570 (1988); T. Huizinga et al, Nature
333:667-669 (1988)). It has been suggested that additional components of
Fc.sub..gamma. receptors may have thus far been missed. Possibly such
components are even more easily lost upon solubilization of the receptors
than are the .beta. and .gamma. subunits of Fc.sub..epsilon. RI (J.-P.
Kinet et al, Biochemistry 24:4117-4124 (1985)). It seems reasonable to
speculate that such hypothetical components would be homologous to .beta.
or .gamma., or both. The availability of genetic probes for the latter
components will now permit an in-depth exploration of this possibility.
The success in expression of IgE binding achieved according to the present
invention has important therapeutic implications. Degranulation of mast
cells and basophils triggered by Fc.sub..epsilon. RI accounts for many of
the symptoms of allergy. Given the high incidence of this disorder, the
discovery of a specific inhibitor of IgE binding is expected to yield
enormous therapeutic benefits. The development of such an inhibitor has
been hampered by the lack of a practical in vitro assay for the binding of
human IgE to the human receptors. For example, a recent assessment of
IgE-derived peptides of their inhibitory capacity had to be determined by
skin-testing (B. Helm et al, Nature 331:180-183 (1988)), a cumbersome and
potentially dangerous procedure.
That the present invention achieves the expression of the transfected
rodent receptor indicates that human Fc.sub..epsilon. RI can be similarly
expressed. Alternatively, since at present only the cDNA coding for the
human .alpha. subunit has been isolated (A. Shimizu et al, Proc. Natl.
Acad. Sci. USA 85:1907-1911 (1988); J. Kochan et al, Nucl. Acids Res.
16:3584 (1988)), it is expected that it can be expressed in
cotransfections with the cDNAs coding for the rodent .beta. and .gamma.
chains.
A comparison between the human and rat .alpha. subunits is set forth in
Table 4 below.
TABLE 4
______________________________________
Comparative Properties of Human and Rat
Alpha Chains
Species
Domain Human Rat % Homology
______________________________________
Extracellular
180 181 49
Transmembrane 21 21 67+
Intracellular 31 20 23
Total 232 222 47*
______________________________________
* Wt ave.
+ Human: WLQFFIPLLVVILFAVDTGLFISTQQQ
Rat: WLQLIFPSLAVILFAVDTGLWFSTHKQ
It may be seen from the above Table that there is an overall homology
between the human and rat alpha chains of about 47%, but an almost 70%
homology in the presumed transmembrane domains. Indeed, when the
transmembrane domains are examined closely, there is a stretch of 10
consecutive residues that are completely identical. This stretch of
consecutive residues is underlined in Table 4.
Since the transmembrane segment is the region of the .alpha. chain that is
most likely to interact with the .beta. and .gamma. chains, it was
expected that the human .alpha. chain would be expressible, if
transfected, along with the rat .beta. and .gamma. chains. This has proved
to be the case as the present inventors have been able to express human
IgE binding by COS cells transfected simultaneously with the human a and
the rat .beta. and .gamma. subunits. It will be advantageous, of course,
to have permanently transfected cell lines and for such lines, one will
want to utilize the human .beta. and .gamma. subunits. The present
inventors are in the process of identifying the coding sequences for these
subunits so that preparing such transfectants will be straightforward.
Thus, with the materials available now, it is already practical to search
for peptide inhibitors of human IgE binding in vitro. To make the assay
suitable for truly mass screening of drugs will require only minor
extensions of the present work.
The genetic work, of course, provides much more than an assay, as important
as the latter may be. Through directed mutation it will, in addition,
allow the development of further information regarding the critical
binding regions. It is expected that, using this information, rational
drug design will become possible. It is further expected that it will be
possible to block the function of the receptor itself, i.e., it will be
possible to interfere with the early biochemical signals that result from
activation of the receptor.
While the invention has been described with respect to certain specific
embodiments, it will be appreciated that many modifications and changes
may be made by those skilled in the art without departing from the spirit
of the invention. It is intended, therefore, by the appended claims to
cover all such modifications and changes as fall within the true spirit
and scope of the invention.
Top